Natural products have historically been the source of most of the microtubule (MT)-targeting small molecules whose properties have allowed them to become useful drugs. That remains true of most but not all of the compounds that we have used in this study. These include the clinically established MT-active drugs colchicine, combretastatin, vincristine, taxol, and others. Almost all such agents were developed first in pre-clinical research that included in vitro studies of the effect of the compounds on polymerization of tubulin to microtubules as well as the effect of such compounds on cell behavior, especially examining the ability of the compounds to disrupt mitosis through effects on the MT arrays that comprise the mitotic spindle. Indeed the ability to cause mitotic arrest in rapidly growing cell cultures in the laboratory is often considered to be an assay of the principal mechanism of these drugs. Despite the fact that we have argued that mitosis is not likely to be a central target for chemotherapy in patient tumors, it is still true that chemotherapy agents can alter mitosis. This is true to some extent for all chemotherapy agents, not only for microtubule-targeting ones. To measure these effects we previously designed a cell-based assay using a cell that contains an artificial chromosome that codes for a fluorescent protein. Exposure of these cells to various chemotherapy agents results in different kinetics of loss of fluorescence due to differing effects on mitotic fidelity and hence chromosome loss. These results allow quantitation of each chemotherapy agents ability to induce chromosome loss. Our results showed that agents differ remarkably in their potency against mitotic fidelity, and that even microtubule-targeting agents differed significantly from each other. Surprisingly, microtubule-stabilizing drugs cause more chromosome loss than do microtubule-destabilizing drugs. These results may allow insight into the mechanisms of mitotic fidelity as well as providing a rationale to favor one drug clinically over another (if they differ in potential for causing chromosome loss). Many chemotherapy agents and other bio-active small molecules act by inducing increased levels of reactive oxygen species. These can oxidize proteins, lipids, and other molecules. Proteins are very abundant in cells and hence are a major target for oxidation. A princlipal product of this reaction is carbonyl groups on proteins. These do not occur normally, so carbonyl accumulation provides a non-reversible read-out of oxidation in a cell. Hydrazines and hydrazides react with carbonyls and provide a group-specific means of assaying the presence of carbonyls on proteins and other molecules. We developed a new fluorescent reagent, a benzocoumarin hydrazine, that provides a covalent fluorescent signal upon reaction with carbonyls. It is cell penetrating, and can be used with live cells as well as in cell extracts or purified proteins or other molecules. We intend to use this assay to expand on our understanding of how chemotherapy agents and other bioactive small molecules induce chromosome loss and other cell changes. In pursuit of newer microtubule-targeting agents with more favorable spectra of actions as well as more facile chemistry, we previously developed a new method for synthesizing variants of polyketides, a class of compounds containing many clinically important natural compounds. We applied our method to the microtubule-targeting natural product drug dictyostatin, demonstrating that this method allowed synthetic extension of particular sites on precursor molecules to produce new variants of dictyostatin that demonstrated significantly different biological activity. In this report year we described the development of new analogs of epothilone, a microtubule-stabilizing drug that is already in clinical use. Epothilones have a number of characteristics that recommend them over the more established paclitaxel (marketed under the name Taxol). They are more water soluble, the synthesis is more facile, and parts of the synthesis can be accomplished by fermentation of suitable organisms. Nonetheless, epothilones are relatively new, and new versions are sought. We describe the synthesis and properties of analogs of 7-deoxy epothilone, establishing the significant regions of this molecule to activity in microtubule-targeted activity. We also describe the expansion of the methyl-extension chemistry we previously described, to produce a version of eothilone that can be easily cross linked to standard amine chemistry to antibodies or other molecules. This provides a straightforward means to couple this highly cytotoxic drug to cancer-specific antibodies and hence allow targeting of the drug to tumors selectively over the normal cells in the patient. These applications are underway. We also pursued the basic structural biology of the tubulin dimer. This knowledge is required to understand how small molecules regulate tubulin and how tubulin binding domains regulate interaction with other proteins. The most basic step in tubulin biology is assembly of the heterodimer, and we previously described studies that used analytical ultracentrifugation as well as fluorescence polarization to show that dimer formation is a reversible, mass-action-driven process. We also examined the role of the tubulin carboxyl terminal tail peptides which mediate interaction of many proteins with tubulin and/or microtubules. We have extended those studies, showing that dimer dissociation and monomer exchange are controlled by equilibrium constants that differ by orders of magnitude between different tubulins. These include tubulins from human cells compared to tubulins from diifferent organisms, including protozoan parasites, and also between tubules from different tissues, which represent different expression levels of tubule isotypes. This work is being prepared for publication. These differences in dynamics are expected to alter cellular morphological plasticity and possibly cell migration.